Valuation of Intellectual Property Assets

The foundation for risk management and financing.

Valuation of Intellectual Property Assets:The Foundation for Risk Management and Financing

Table of Contents

Page

Nature of intellectual property...................................................................................................1

Interrelationship of intellectual property toassets of a business enterprise ............................................................................................4

What do banks and insurance companieslook for in an IP valuation?................................................................................................5

Valuation methodology .............................................................................................................8

Economic contributions of intellectualproperty - a market perspective ........................................................................................22

Conclusion .....................................................................................................................27

- 1 –

Valuation of Intellectual Property Assets

The Foundation for Risk Management and Financing 1

Zareer Pavri, Vice President, PricewaterhouseCoopers

My purpose is to outline a methodology for establishing the fair market value (defined later) of intellectualproperty that may be associated with a going-concern business enterprise. I emphasize "outline", since adetailed presentation of this topic could be the subject of yet another textbook. It is, however, still possibleto achieve my objective in this format by identifying the elements of intellectual property and providing thebasic concepts for valuing them.

Within this context, I intend to:

• identify the nature of some of the more common or well-known forms of intellectual property;

• discuss the normal relationship of these forms of intellectual property to the monetary and tangibleassets of a business enterprise from an investor's perspective; and then

• discuss the methodology for valuing them.

Nature of intellectual property

1 Presented at the INSIGHT conference: “Negotiating License Agreements: Maximize the

Value of Your Intellectual Property Assets”, Toronto, April 29 – 30, 1999.

Intellectual property acquires its essential characteristics, from which value emanates, from the legalsystem. By this, I mean that the law gives rights to people who create things that embody new ideas orways of expressing ideas, and to those who use certain marks to distinguish their product or service. It isthis unique characteristic of intellectual property (i.e. legal protection) that causes it to be a subset ofintangible assets of a business enterprise. The property may have resulted from arduous and costlyresearch or simply by fortuitous discovery. Nevertheless, intellectual property may contribute significantlyto the earning power of an enterprise of which they are a part. Intellectual property often requires hugecapital outlay to create. Once established, however, many elements of intellectual property of a businessare unique, commanding and a driving force, yet fragile.

- 2 –

The following listing provides examples of those intangible assets that are created by the business andthose that exist under the protection of law:

Created by business Exist under legal protection*

• an assembled trainedworkforce

• advertising programs• distributor networks• training materials• customer loyalty• supplier contacts• management depth and

experience• subscriber base• goodwill

• patents (20 years)• trademarks (15 years +

renewals)• copyrights (50 years)• industrial designs (5 years

+ 1 renewal)• trade-secrets and know-how;

collectively "proprietarytechnology" (contractperiod)

* Years in parenthesis indicate legal protection period.

The list is not meant to be all inclusive and certain industries may have other valuable intangible assets. To the extent that computer software can be subject to patent or copyright protection (the latter being moreeasily obtainable), it can also be included as intellectual property.

A brief discussion of the components of intellectual property is provided below:

Patents -

Patents protect function: the way the parts of a machine or product functionally interact or thefunctional steps of a method. A patent gives to the inventor of a new and useful product ormethod, the exclusive right to make and sell product, or to use that process in Canada for a periodof 20 years (old Act, 17 years). The 20-year period begins from the date the patent application isfiled. The monopoly is not enforceable until the patent is granted, usually a couple of years later.

Trademarks -

A trademark is a word(s), logo, design, number(s) that is used to distinguish a person's product orservice from those of others. It can be the brand name of the product or service, but it cannot be ageneric name of the product or service itself. A trademark registration in Canada is granted by thefederal government only after an application for registration has been filed. A trademarkregistration gives the owner the right to use the trademark in association with the wares and

- 3 –

services specified in the registration. Trademarks protect only the device or symbol attached to thegoods or services, not the goods or services themselves; the latter are open to the world. Thetrademark's owner is entitled only to prevent it from being used to make purchasers believe theyare buying his or her goods when, in fact, they are buying those of a rival. The registrationprovides such protection for a period of 15 years, with repeated renewals on payment of a fee solong as the use of the trademark continues.

Prior to 1993, all a trademark owner had to do to preserve his/her mark was to register thelicensees in the Trademarks Office. In 1993, the Trademarks Act was amended to require alllicensees to be licensed by or with the authority of the owner; and under the license, the ownermust have direct or indirect control over the “character or quality” of a licensee’s wares orservices.

A “trade name” (or trading style) is quite different from a trademark. A trade name is the nameunder which any business is carried on, be it a corporation, partnership or individual. A tradename is attached to a business. It cannot exist "in vacuo"; that is, independent of the businessenterprise which bears the trade name. A trademark differs from a trade name in that the former isused in association with vendible commodities or services, while the latter is more properly linkedto an enterprise's residual value, also referred to as goodwill.

A company's success in establishing a recognized trademark depends to a large degree, of course,on its reputation for quality products or services. In most cases, however, trademarks aredeveloped and maintained primarily through extensive and costly advertising.

Copyrights -

Copyright protects the form of expression of an idea, but not the idea itself. Copyright protectsworks of art or sculpture as well as literary and musical works. Copyright arises automaticallywhen an "original work" is created; no registration is necessary. The copyright owner has theexclusive right during the author's or composer's lifetime, and in most cases for 50 years after hisor her death, to grant permission to copy or perform the work. Copyright and Industrial Designs(see below) are mutually exclusive, that is to say, a design capable of being registered as anIndustrial Design cannot be protected by copyright.

In 1988, the Copyright Act was amended to include computer programs within the definition ofliterary works. In addition, data bases are generally protected by copyright as compilations andthrough common law as proprietary information.

Industrial Designs -

An Industrial Design protects the form of industrial products such as their ornamental aspects andshape, not their functional aspects. An Industrial Design lasts for 5 years and is renewable foranother 5-year term.

- 4 –

Trade Secrets and "know-how" -

Trade secrets and confidential information are data that are collected or created by an entity for itsown use. They may for some reason be disclosed to another who is not obliged to disclose it toany one else. The obligation is evidenced by a contract between the two parties. The parties agreeto do something (exchange information, yet retain its confidentiality to the outside world) or not dosomething (not to use information unless specifically authorized to do so).

Interrelationship of intellectual property to assets of a business enterprise

A business enterprise has three basic components: working capital as represented by current assets lesscurrent liabilities, plant assets, and intangible assets. These components are funded by a combination ofdebt and equity. Shown below is a balance sheet as it would appear to an investor or someone concernedwith the value of a business.

WORKING CAPITAL

PROPERTY, PLANT & EQUIPMENT

OTHER ASSETS (if any)

IDENTIFIABLE INTANGIBLE ASSETS

UNIDENTIFIABLE INTANGIBLE ASSETS

In a valuation context, the term "working capital" means the average or typical working capital requirementof the business and is not necessarily represented by the actual excess of current assets over currentliabilities at any specific date. 'Property, plant and equipment' would represent the current market value ofthose assets as well as of "other assets", if any.

Intangible assets represent all the elements of a business enterprise that exist after monetary (workingcapital) and tangible (plant, etc.) assets are identified. They are the elements, after working capital andfixed assets that make the enterprise 'tick' and contribute to the enterprise's earning power. Their existence,and particularly their value, is dependent on the presence, or expectation, of earnings. They appear last inthe development of a business and disappear first on its demise.

Identifiable intangible assets - typically patents, trademarks or brand names, copyrights, etc. - are elementsgenerally classified as intellectual property. Any residual value that may be left after accounting for the

- 5 –

identifiable tangible and identifiable intangible assets is attributed to unidentifiable intangibles of whichgoodwill is the most common. It would not be unusual to find that a company may be successful andcontinuing but not possess goodwill as a separate intangible asset. Continued patronage (a commondefinition of goodwill) can sometimes be completely attributed to identifiable intangible assets such astrademark recognizability or patented product attributes. Allocation of the remaining amount of theenterprise value is then not complicated by allocating an amount to goodwill. In situations where goodwillmay still be considered to exist, it may indeed be very difficult if not impossible to precisely segregateintellectual property value from goodwill.

What do Banks and Insurance companies look for in an IP valuation?

From the Canadian banks’ and insurance companies’ perspectives, this whole area of IP valuation isrelatively new and continually evolving. Consequently, there is no general agreement within theirrespective circles as to an accepted valuation methodology. Having said that, the following observationshighlight how Canadian banks (from a lending perspective) and insurance companies (from a first partyliability perspective) view intellectual property assets and their value:

Banks –

• Faced with the rise of the knowledge-based economy, Canadian banks have re-calibrated their lendingcriteria and their relationships with knowledge-based firms.

• Today, all the major Canadian banks have setup and operate specialized groups with innovativeknowledge-based industry strategies. Information networks and alliances with non-bankingprofessionals are helping these specialized groups to become more effective at the early stage ofdevelopment of emerging companies.

• Banks typically do not lend money to finance development stage companies or emerging technology. They get involved when there is market acceptance of the technology-based product that eventuallytranslates into inventory and accounts receivable which assets then form the basis of grantingtraditional operating lines of credit.

• With respect to established consumer brands, the banks have begun to move away from a purelybalance-sheet analysis into looking at the cash flow generating capability of the business. [Here, cashflow is generally defined as operating income after adding back non-cash items such as depreciationand amortization, less sustaining capital expenditures and working capital changes, if business growthis rapid. Some even factor in sustaining ‘R&D’ if it is segregable from developmental ‘R&D’].

• Increasingly, established brands are viewed as being important factors in assessing overall credit. Strong brands reduce cash flow volatility; lower volatility commands higher multiples (all other factorsremaining unchanged), which in turn results in higher lending values.

• Canadian banks recognize established brands as valuable assets, but they are not comfortable at

- 6 –

quantifying brand values, per se; nor do they need to, according to at least one banker. The overalllending objective is whether there is adequate, stable cash flow from the business (owning the brand)to service and repay the loan within the stipulated time period. Perceptions of brand values aretherefore reduced to a “floor” value that will provide the lender enough cushion to justify the corporatelending risk.

• Valuable brands are seen by banks as adding to the quality of the lending opportunity. Banks try to“control” valuable brands during the tenure of the loan by various methods such as incorporatingnegative pledges (whereby the borrower pledges not to sell, hypothecate, or otherwise diminish thevalue of the brand) in loan agreements and the inclusion of established brands in GSA’s (generalsecurity agreements).

• Gradually, and where circumstances justify, Canadian banks have begun to include identifiableintangible assets (trademarks, patents, brands) recorded on a company’s balance sheet as part of “nettangible worth” when stipulating debt: equity covenants in loan agreements.

• In contrast, lending in the U.S. and U.K. using brands as collateral is not uncommon (e.g. the RJRNabsico LBO, Sara Lee financial engineering and Border Cos.’ special purpose vehicle-relatedfinancing). Moreover, in the case of companies like UK-based Ranks Hovis McDougall, brandvaluation is built into borrowing covenants such that it is considered an asset for borrowing purposes.

• There are at least three reasons, I believe, for the differences in the level of lending activity in thisrather specialized field and these are:

(i) given the thinness of the Canadian market and paucity of world-class brand names, theredoes not appear to be the appetite amongst Canadian banks to undertake such financing;

(ii) the overall conservatism of Canadian banks coupled with a reluctance (at least at thepresent time) to accept generally used IP valuation methodologies as standard; and

(iii) issues regarding title to brands or rights to brand usage, given the preponderance offoreign subsidiaries in Canada.

In what was envisioned two years ago as a new market for so-called asset-backed bonds – bonds backed bymusic royalties, or royalty bonds – the banking community watched gleefully as investors scrambled tosubscribe to a US $55 million bond issue backed by the future earnings of David Bowie’s (one of theworld’s top pop artists) recordings made prior to 1990. At the time, Prudential Insurance Companysnapped up all the bonds providing it with a 7.9% return on its investment over ten years, and allowingDavid Bowie to collect the US $55 million up front.

Yet two years later, the development of the market for securitizing music royalties has been significantlylower than expected. One explanation is the impact of the changing economic climate on the financial

- 7 –

markets and general investment trends. Investors have become warier of experimenting with new types ofinvestments and have also shown a marked preference for liquid assets. Royalty bonds are illiquid becausethere is no secondary trading on them.

Another explanation for the less than stellar development of music royalty bonds is that only a handful ofrecording artists actually own their own master types (the rights to the original recordings) as David Bowiedoes. Most others, like Rod Stewart, ceded control of their mastertapes to their record companies in theirrecording contracts and receive royalties simply on the income generated by them.

Insurance Companies -

• Protecting a product’s brand name is a major concern for food, beverage, pharmaceutical, cosmeticand tobacco companies. Ingestible or topical products are susceptible to contamination bothaccidentally and maliciously. Infections from improperly heated or sealed canned goods or cookedfoods, mislabeled drugs, error in mixing cosmetic ingredients as well as intentional contaminations canall have a devastating impact on the consumer confidence in the affected product. The tampering ofSudafed and the Tylenol poisonings caused major financial loss to the companies involved in theseincidents. Even “bogus” tamperings such as the Diet Pepsi/Syringe “scare” in 1993 can affectconsumer confidence in a product’s safety. As a matter of interest, Pepsi reportedly went to greatlengths and expense to prove that a syringe could not possibly be inserted into a Diet Pepsi can bygoing live and demonstrating that it could not be done.

• Another disturbing trend is afoot: the use of the World Wide Web by social activists, pranksters andthe just plain disgruntled to alter and damage a company’s trademarked image with negative publicity. Whatever form the negative publicity takes, it is almost instantly available to the growing millions ofpeople, more of whom are consumer-oriented, who log on to the Internet.

• Recognizing the catastrophic impact that incidents of this type can have on a business, there is now ahandful of insurance companies (as far as I have been able to determine) which offer a level offinancial protection against product contamination. Among these, AIG, an off-shore insurance group,has been a pioneer in offering coverage against accidental product contamination, and since 1983/84,has offered an insurance product to cover malicious product contamination. Contaminated ProductInsurance or CPI is a first party liability coverage aimed at two perils:

- malicious product tampering which involves deliberate tampering of a product making it unfit forconsumption resulting in damage to the brand name, if any, loss of profits, drop in stock price etc.

- accidental contamination which generally occurs in the manufacturing process with the productgoing to market, not performing well or selling at all, thereby again resulting in damage to thebrand name, loss of profits, etc.

• Product tampering appears to be a phenomenon of the developed countries. Potential for the CPIproduct exists primarily in the U.S., U.K., France, Germany and Australia. If at all there is a market in

- 8 –

Canada, it is relatively insignificant at the present time.

• On a worldwide basis, the CPI market is considered quite small - estimated annual premiums ofUS$100 million plus.

• There is no scientific valuation methodology used by insurers in setting brand values for CPI coveragepurposes.

• Several factors are, however, considered among which are:

- brand product(s) description and profile (the higher the profile, the greater the risk)- sales volume and revenue (historical and projected) and its geographical dispersion- reliance on one brand or several (dependence on several brands diffuses risk of loss)- manufacturing and packaging processes (easy to tamper?)- the effect that ingredients used or end product could have on special interest groups- disgruntled employees, if any- geographical dispersement of operations and market (U.S. considered more litigious than most other countries)- internal security

• According to AIG, brand value coverage is formulated, initially, from a base value for brands in asimilar industry premised upon the insurer’s cumulative experience. Subjective adjustments to suchbase value are then made upon a review of the aforementioned specific factors.

• CPI risk is controlled by the insurer placing an insurable limit. In the case of malicious productcontamination, AIG places a US$70 million limit for each brand name product, with a deductible of1% of the policy limit and annual premiums that are quite “reasonable”, according to a senior official. Because of the broad coverage (of products and events that may cause contamination) relating toaccidental product contamination, insurable limits are much lower (around US$10 million), co-insurance (generally 25%) is a pre-requisite and premiums are accordingly relatively high in relation tothe insurable limit.

• Limiting the liability is, however, of little use to owners of high value brands (those valued in thehundreds of millions or even billions); the only alternative being self-insurance in such cases.

Valuation methodology

Given that value is generally defined as the present value of future benefits to be derived by the owner ofproperty, a valuation needs to quantify the future benefits and then calculate present value.

In this paper references to value are to fair market value which is generally defined as the highest priceexpressed in terms of money or money's worth that would prevail in an open and unrestricted market

- 9 –

between a willing and informed buyer and a willing and informed seller, each acting at arm's length andunder no compulsion to transact.

Some basic valuation approaches should be understood before any attempt is made to address the ratherspecialized field of intellectual property valuation. Because these assets can be and are often transferredindependently of the business enterprise that owns them, I will introduce one or more of the morecommonly used techniques of valuing intellectual property per se, as distinct from the valuation of thebusiness enterprise as a whole.

Having said that, the three accepted valuation approaches used to value a business enterprise are also thoseused to value intellectual property, with varying degrees of applicability. The three approaches are: cost,income and market.

The cost approach seeks to measure the future benefits of ownership by quantifying the amount of moneythat would be required to replace the future service capability of the subject intellectual property. This isknown as the cost of replacement. The underlying assumption is that the cost of new property iscommensurate with the economic value of the usage that the property can provide during its life. Absentmarket aberrations, the cost of a brand new property is its fair market value. Since one is seldom calledupon to value a brand new property, the application of the cost approach always brings with it thecomplexity of quantifying the reduction from (brand new) value due to depreciation. In value terms,depreciation can result from physical use, functional obsolescence and economic obsolescence and aproper reflection of all these three forms of depreciation are required in order to arrive at value using thecost approach. The cost approach does not directly consider the amount of economic benefits that can beachieved nor the time period over which they might continue. Economic benefits are simply assumed toexist in sufficient quantity and duration to justify developmental expenditures. Critics of this approachalso point out that the cost for a successful venture must reflect the cost of failures.

This is particularly true in the pharmaceutical industry where research and development costs for failuresmay run more than five times the cost of research and development successes. As such the cost approachhas limited application in valuing those forms of intellectual property that are capable of being transferredindependently of the business of which they are a part.

The income approach, in contrast with the cost approach, focuses on the consideration of the income-producing capability of the intellectual property. The underlying assumption here is that value is measuredby the present value of the net economic benefit (cash receipts less cash outlays) over the life of the asset. Where conditions are not conducive to deriving economic benefit (or profit), it is difficult to ascribe anyvalue (taken to mean fair market value defined above) to intellectual property regardless of the indicationsof the cost approach. The income approach is best suited for the valuation of intellectual property such aspatents, trademarks and copyrights.

The market approach is generally the most direct and most easily understood valuation approach. Itreflects the value obtained as a result of a consensus of what others in the market place have judged it tobe. For this to occur, there must be an active public market and an exchange of comparable properties.

- 10 –

Where the optimal market does not exist, application of this method becomes more judgmental to the pointthat it becomes a less reliable measure of value. This method is seldom used in the valuation of intellectualproperty primarily because there is rarely an active market in which public information (as to price andinherent comparability characteristics) is readily available.

The following table summarizes the various suggested methodologies by type of intellectual property. Themethodologies are denoted (1) for primary method and (2) for secondary method or 'sanity' check to beused when relevant information is available.

MethodologyIntellectual Property

Type Priority

Patents- active

- inactive (but potential for future use)

Trademarks

Copyrights

Secret processes, 'Know-how',engineering specs, etc.

Computer software- patent or copyright- proprietary/dedicated

for 'in-house' use

income (1)cost (2)

cost

income (1)cost (2)

income

cost

income

cost

As a general observation, the cost approach is used as a primary method (i)when it is not feasible to projectearnings for the intellectual property, (ii) it is not the type of asset that can readily be transferred to a thirdparty separate from the organization within which it resides, or (iii) where the intellectual property (e.g.computer software) is developed for “in-house” or proprietary use and not for resale. While a detaileddiscussion of the various cost estimation methods to value intellectual property such as computer softwarecould be a separate topic for a paper in itself, suffice it to say that several cost estimation methods exist:expert judgement, estimation by analogy, Parkinson estimation, etc. However, of the alternative softwarecost estimation methods, the various algorithmic models, of which the constructive cost model

- 11 –

(“COCOMO”) is one, are considered to be the more reliable and supportable models.

I will now expand a bit on the more commonly used income approach. Within the income approach, thereare various methods I shall be discussing or referring to such as:

• residual method• excess income• relief from royalty method• capitalization of profit margin differential

References to "introductory and maintenance brand spending differential" and "carry over benefits of pastadvertising costs" methods, both of which are cost-based rather than income approaches, are mentionedhere to provide a more complete picture of available methods.

It should be noted at the outset that this discussion concerns forms of intellectual property (i.e. patents,trademarks, etc.) that are part of a business enterprise. They are therefore capable of producing income forthat enterprise and their worth is predicated on that capability. Therefore, whatever the intellectualproperty may be, we must look toward some product or service with which it is associated. It is thatproduct or service, converted into money in the marketplace, that is the source of economic benefits bywhich value can be measured. It is therefore necessary to link the intellectual property being valued with aproduct or service either existing, or contemplated. Without this linkage, the asset can have no value andno economic life (see comments below). In order to properly assess the economic life remaining at thepoint in time when a valuation is being done, the valuator has to determine where in the life cycle is theproduct with which the intellectual property is associated.

The three essential ingredients of the income approach are:

(a) The amount of the income stream that can be generated by the product or servicewith which the intellectual property is associated.

(b) An estimation as to the duration of the income stream, i.e. the economic life of theproduct or service with which the intellectual property is associated.

(c) An assumption as to the risk associated with the realization of the forecasted income. Risk reflects all the business, economic and regulatory conditions associated withemploying the property and achieving the prospective earnings.

Attention to forecasting is important in the valuation of intellectual property because value is solelypredicated on the anticipation of future economic benefits. When forecasting revenues and expenses,several factors come into play such as economic and macroeconomic variables, political risks, rawmaterials availability, labour availability, capacity, population trends and demographics etc. There are

- 12 –

many methods of forecasting: time series forecasting, exponential forecasting, the use of historical andcurrent data to project into the future. Judgmental methods (relying heavily upon the informed opinion ofleading experts) are used to forecast prospects for new products for which historical data do not exist. Afull discussion of the forecasting methods is beyond the scope of this paper. Suffice it to say thatforecasting is a key element in intellectual property valuation.

Value and economic life have a very close relationship. Economic life could be described as the periodduring which it is profitable to use an asset. Economic life ends when (1) it is no longer profitable to usean asset, or (2) when it is more profitable to use another asset. This is quite different from the "servicelife" of a tangible asset which is the period from its installation to the date of its retirement or from "legallife" in the case of intellectual property which ends on the expiry of legal protection, irrespective of theearnings capability during the period of the product or service with which the intellectual property isassociated.

In general, the process of estimating economic life of an intellectual property is one of identifying all of thefactors that bear on economic life in a given situation and then making a judgement as to which of themindicates the shortest life. Other than in exceptional circumstances economic life of an intellectualproperty cannot extend beyond its legal life.

Outlined below are some of the factors one would consider in estimating economic life of the followingintellectual property:Patents -

• loss of supply or price escalation in raw material that could render process uneconomic

• increase in energy costs that would render process uneconomic

• legislation relative to environmental concerns affecting the process or product

• the probability of a competitor designing around the protected process

• development of a superior process that would replace the existing one

• challenges of patent validity brought about by competitors

Most difficult is the estimation of economic life of embryonic technology or unproven product. In suchcases, an "educated guess" is often the only solution, recognizing that the margin for error may besignificant.

Trademarks -

One school of thought supports the notion that trademark rights have unlimited economic lives,

- 13 –

since they exist as long as they are used and maintained. Another school of thought offers forconsideration the idea that because trademarks are and, in fact, can be maintained in "100 percentcondition" or even improved with continual advertising etc., a trademark is, at any specificmoment, the product of investments of the past. If future investments to maintain it (in both labourand monetary terms) were to cease, it is clear that a mark would die. While this may be an extremeassumption, it is submitted that future investments can replace those made in the past, andtherefore the value of a trademark at a specific point in time will diminish, its place then beingtaken up by the new investment.

For practical purposes, corporate trademark valuations cover a period of 20 years for present valuediscounting purposes. Product trademarks or product brand names typically have shortereconomic lives with factors similar to those for patents determining their economic lives.

Copyrights -

Copyrights enjoy a long legal life; in economic terms, however, their lives are much shorter andmost often their benefits are not evenly accrued over that shorter life. There is such a variety ofcopyrighted work that it would be impossible to make statements across the board as to what theireconomic life might typically be. Economic life in this case would depend on the type of work andthe manner in which it can be exploited. Some of the factors that merit attention in estimating theeconomic life of copyrights include:

(a) the breadth of exploitation. Cartoon characters are a good example. It is commonfor such characters appearing in books to be exploited in a variety of media aswell as ancillary products such as T-shirts;

(b) versatility. The more versatile the copyright the greater the opportunities forexploitation;

(c) timelessness. For example, a motion picture on the Middle East War would havehad its run by now, while Disney motion pictures made decades ago, areplaying to a new generation of children. It should be noted that Disneypreserves the "timelessness" of its movies by intentionally withdrawing certainmovies and all direct and indirect rights to those movies (e.g. "The LittleMermaid") for 12 to 15 years, awaiting a new generation of audience.

Trade secrets and "know how" -

Most of the considerations used to estimate economic lives of patents apply here as well, with theexception that there is no statutory limit to trade secrets protection. Some additional uniqueconsiderations include:

- 14 –

(a) transferability of the trade secrets. By this is meant the extent to which suchinformation is reduced to writing or other transferable form. Certain skills("know-how") such as those of a writer, musician or surgeon can be extremelyvaluable know-how, but largely non-transferable and therefore of nocommercial value;

(b) care with which the confidentiality of information is preserved or protected;

(c) versatility of the know-how can enhance its economic life; in other words, can it bere-deployed if there is a change in the market? If it can, then economic lifewill be enhanced.

Valuation text books refer to two commonly used income methods to value intangibles: (1) the residualtechnique and (2) the excess income method. Their application in valuing a particular identifiableintangible asset per se, such as a patent or a brand is questionable, in my view.

For instance, the residual method, if at all it results in a reasonably precise value of a patent, does so in anindirect way. The value so arrived at is dependent first on the value of the enterprise as a whole, secondly,on the allocation of value to the tangible assets thus leaving a block of value allocable to intangible assetsand finally, on the allocation of this block to the various forms of intangible assets that may exist in anenterprise. It is the third step which is most subjective and one that can often test the realm of reasonedjudgement. Any value left over is then attributed to the subject intellectual property. Any error in aprevious allocation will directly impact the value of the subject property. To be fair, however, this methodin certain circumstances (refer to case study later in this paper), does provide an indication of a market-driven royalty rate for a bundle of intellectual property.

The excess income method employs a technique known as the dual capitalization method where one rate ofreturn is applied to tangible assets and another (normally higher) rate to intangible assets in order to reflectthe inherently higher risk of such assets. This method does not in and of itself enable a valuator tosegregate value between identifiable and unidentifiable intangibles.

Segregation of value as between identifiable and unidentifiable intangible assets is required becauseintellectual property (an identifiable intangible) particularly if valuable is seldom sold to but is oftenlicensed for use by third parties in return for a royalty fee. As I have explained later, value and the royaltyfee that such an asset can command are closely inter-linked. Moreover, in the purchase of assets (asopposed to shares) of a company, where the buyer is required to allocate the purchase price to tangible,identifiable intangible and unidentifiable intangible assets, it is not only desirable but also a requirement todetermine the fair market value of those intellectual properties such as patents, licenses, franchises,copyrights, etc. that fall into a specified CCA class for tax purposes.

Given the limitations of the two aforementioned income-based methods, one of the more generallyaccepted methods, and one that I have used, to value a patent (and for that matter a trademark or productbrand name) per se is the "relief from royalty" method. The concept here is that, by owning a patent, a

- 15 –

company is relieved of the necessity of having to pay royalties for exclusive/non-exclusive patent rights. Itfollows, therefore, that anyone wanting to obtain the right to the patent or trademark would have to enterinto a business arrangement with the original owner. Such arrangements akin to licensing agreements,usually entail a royalty payment, typically expressed as a percentage of net product sales.

A central issue in adopting the "relief from royalty" method is the determination of an appropriate royaltyrate. The simple solution would be to locate an exact comparable licensing transaction between unrelatedparties. Unfortunately, like fingerprints, no two license agreements are exactly the same. In the absence ofan exact comparable, justification for an appropriate royalty rate often defaults to similar transactions or"inexact" comparables (also referred to as comparable adjustable transactions) or to the broader standard of"prevailing rates" in the industry.

Royalty rates, however, cannot be evaluated in a vacuum. Arm's length licensors and licensees negotiateroyalties within a dynamic matrix of strategic, economic and legal considerations. Each term and conditionin a license agreement may shift risk from one party to the other and, therefore, should be considered indetermining the appropriate royalty rate or range thereof.

It follows that one of the first steps in adopting this method is to carry out a functional analysis. Thisincludes identification of the relevant intellectual property which is the subject of royalty rate finding,determination of economic "value added", comparing the economically significant activities undertaken orto be undertaken by the licensor and licensee and determination of the nature of the intangible (i.e."routine" or "non routine") from the licensee's perspective. As the term suggests, a "non-routine"intangible property, such as a patent for a unique chemical compound, is central to the conduct of abusiness and without which the business activity could not be conducted. Its uniqueness is of significantvalue to a licensee and hence would command a much higher royalty than a routine or substitutableintangible. The functional analysis would also address many functions performed by a licensor and alicensee - a "who does what" list in terms of R & D; product design and engineering; manufacturing andprocess engineering; product fabrication; marketing, advertising and distribution; transporting andwarehousing etc. In short, a functional analysis is designed to establish the facts on which the subsequentdetermination of an appropriate royalty rate is based.

There are several factors that affect royalty rates and these factors may have varying degrees of importancedepending on whether they are viewed from a licensor's perspective or a licensee's perspective. A recentU.S. survey of licensing out (licensor perspective) and licensing-in (licensee perspective) agreementsranked the importance of these factors as shown below:

Factors

ModeratelyImportant Important

VeryImportant

Licensor • Enforcement • Utility/ • Protection

- 16 –

burden• Foreign vs

domesticpartner

• Convoyed salesopportunities

advantages• Commercial

success• Refinement• Competition• License

duration• Minimum

royalties• Support/

training• Licensor

commitment

• Exclusivity

Licensee • Competition• Licensor

commitment• Enforcement

burden• Foreign vs.

domesticpartner

• Convoyed salesopportunities

• Commercialsuccess

• Refinement• License

duration• Minimum

Royalties• Support/

Training

• Protection• Exclusivity• Utility/

advantages

The survey confirms that intellectual property protection and exclusivity are very important factors insetting royalty rates in licensing agreements. A 20% to 50% premium has been discussed as a reasonableaverage for exclusivity of license. As much as 300% premium for exclusivity has been reported in thepharmaceutical field.

The process of developing and bringing a product to market can have a direct bearing on royalty ratedetermination. A good example of this would be the pharmaceutical industry where bringing a product tomarket often consumes considerable time and capital resources in obtaining regulatory approval, withoutwhich these products could not be sold. In contrast, in the automotive and computer industry, there is lessgovernment regulation and generally a short time period required to bring a product to market. Therefore,a company that wishes to license a government-approved pharmaceutical product most likely will bewilling to pay more for technology than if it wished to license an automotive product, even if revenue andprofit expectations were similar for the two technologies.

Royalty rates can also vary depending on the state of 'market readiness' of technology. For instance, in thepharmaceutical industry one may find that licensees are willing to pay up to 2% royalties for process,formulation, or software technology, 2% to 5% for pre-clinical compounds, 5% to 10% for early stage

- 17 –

clinical compounds, and 10% to 15% for late stage clinical compounds.Also worth noting is that, I have seldom, if ever, come across an “untainted” or pure royalty rate, that is, anarm’s length rate devoid of peripheral attributes such as pre-determined time period, graduating/decliningscales (royalty rate percentages and/or sales levels), lump-sum/up-front payments, milestone payments,minimum or guaranteed royalty payments, outright stock purchases of or options to purchase stock in thelicensee or licensor, etc.; all of which have varying degrees of impact in establishing the royalty ratestipulated in a license agreement.

There are various sources which may guide a valuator in selecting a benchmark royalty rate:

• license agreement covering a similar patent or trademark granted by the licensor owning the subjectproperty to a third party, or failing that to its foreign affiliates;

• industry associations some of whom track rates in licensing agreements;

• trademark and patent lawyers who frequently draw up licensing agreements;• licensing agreements filed as part of documents in a legal dispute that has gone before a competent

authority;

• BNA (Bureau of National Affairs Inc.) a U.S. publication house puts out Special Reports on transferpricing issues which sometimes contain information relevant to this subject;

• news articles and magazines such as Financial World which provides an annual survey of some top-value brand names; and

• one's own experience as a valuator.

To assist in identifying market-comparable royalty rates, PricewaterhouseCoopers Canada is in thecontinual process of developing and maintaining a royalty rate database that at the present time comprisesjust under 330 transactions. The information in this database is primarily from U.S. licensing economicjournals, Canadian IPO’s and prospectuses and various other public sources.

In order to provide you with a profile of this database, the following chart provides an incidence ofoccurrence of royalty rates by range and by selected industry categories. It would be imprudent to placetoo much emphasis on the information in the chart without independently evaluating the economics of eachlicensing opportunity.

- 18 –

Royalty Rates

Industry0%-

<5%

5%-

<10%

10%-

<15%

15%-

<20%

20%-

<25%

25%-

<30%

30% &

over

Average

%

Apparel 60% 40% 4.00%

Automotive/parts 83% 17% 1.92%

Biotechnology 50% 17% 25% 4% 6.96%

Communications 44% 11% 11% 11% 22% 8.18%

Computers 50% 25% 10.75%

Computer Software 18% 27% 18% 9% 9% 18% 12.00%

Consumer Products 63% 25% 13% 4.34%

Copyrights and Trademarks 40% 40% 10.60%

Distribution 40% 60% 5.15%

Drugs 13% 88% 6.84%

Electronics 30% 50% 10% 10% 6.60%

Entertainment 18% 9% 14% 27% 18% 15.50%

Food 45% 45% 9% 5.47%

Hospitality/Leisure 29% 71% 4.43%

Industrial Products 40% 42% 13% 2% 2% 6.43%

Medical/Medical Products 32% 33% 19% 8% 3% 2% 4% 8.33%

Services 50% 17% 33% 5.79%

Toys & Games 25% 25% 25% 25% 13.13%

- 19 –

Some comments and observations on the database:

• A total of 329 transactions have been included in the database to date. Out of these, 33 transactions(or, 10%) did not disclose royalty rates and 41 transactions (or, about 12.5%) had royalty ratesexpressed as dollars per unit rather than percentage of sales. The above chart therefore comprises 255transactions.

• The most frequently occurring royalty rate percentage in this database is 10% with 37 appearances; thesecond place to 5% with 34 appearances and the third to 3% with 22 appearances. The incidence ofoccurrence of royalty rates in this database could change over time as the population of transactionsreviewed increases or the industry mix changes.

• In the biotechnology sector, an amendment to a license agreement to remove exclusivity reduced theroyalty rate by 27%.

• While the greatest incidence of royalty rate percentages in the biotechnology sector was in the broadband of 0% to <10%, perhaps on the low side by conventional wisdom, most of the transactionsinvolved up-front license fees, acquisitions of or options to acquire stock in licensee companies andembryonic stage of technology, all of which tend to lower the royalty rate that would otherwise beapplicable.

• Initial payments and royalty rates are tending to increase in the drug industry mainly because of thepaucity of new drug discoveries, and the time and funds required to bring new drugs to market.

• In computer software, educational/virtual realty-related software commands higher royalty rates thanindustry norms of 1% to 7%.

A 1998 U.S. appeals court case (Nestle Holdings Inc. et al. vs Commissioner No. 96-4158) challenged theuse of the relief-from-royalty method in valuing trademarks in the context of a sale. The court found thatthis method “necessarily undervalues trademarks”. It went on to say that while the relief-from-royaltymethod is appropriate to estimate damages from the misuse of a patent or trademark, the value of atrademark to its owner (or for that matter, an acquiror who would then own it) is greater than the relief theowner receives from simply not having to pay for the use of the trademark as a licensee (i.e. the principalconcept underlying this method). Ownership of a trademark, the court continued, was more valuable thana license because ownership carries with it the power and incentive both to put the mark to its most valueduse and to increase its value. A licensee cannot put the mark to uses beyond the temporal or otherlimitations of a license and has no reason to take steps to increase the value of the trademark where theincreased value will be realized by the owner. No alternative method of determining the value of atrademark in a sale context was suggested by the appeals court.

A Canadian case involving Tele-Direct’s “Yellow Page Mart” also recognized the value of ownership of a

- 20 –

trademark as distinct from a license.

In a way, these cases highlight the difference between the value of a trademark or patent for licensing – outpurposes (i.e. from an owner/licensor’s perspective) and value for licensing-in purposes (i.e. from alicensee’s perspective). I am of the view that the relief-from-royalty method is valid in valuing thetrademark for licensing-in purposes, but recognition for ownership should be factored in a valuation of atrademark in a sale context under appropriate circumstances.

These circumstances may include, for instance, a conscious plan by the owner to grant few and onlyexclusive licenses, as opposed to one whose primary business is to license-out non-exclusive licenses. Even so, there may still be some value for control in the latter case, although it may not be very significant.

Where circumstances do warrant an upward adjustment for control, this could be done in any one of thefollowing ways:

1. increasing the cash flows to take into account additional rights, line or brand extension to otherproducts and/or industries, etc.;

2. increasing the royalty rate from that ordinarily applicable in a purely licensing-out arrangement; or

3. adding a control premium to the present value of the trademark otherwise determined under alicensing-out arrangement using the royalty relief method.

The quantum of this control premium is a matter of judgement and particular fact situation. Generallyspeaking, in a corporate takeover the premium for control (over the trading stock price) is thought to be inthe range of 10% to 15% for a stand-alone company, absent any special purchaser premium. Thispercentage range, with exceptions falling on either side, could be a useful proxy for an upward adjustmentwhen valuing a trademark or patent in a sale context.

Two income-based methods in arriving at a reasonable royalty rate are: (1) the Analytical Method and (2)the 25% “Rate-of-Thumb” method.

Briefly, the analytical method compares the profits of a brand name or patented product to the profits of ageneric product. For example, consider that Black & Decker’s “Snakelight” flashlight, a brand nameproduct, has sales of $250,000,000 and an operating profit (typically, before corporate head officeexpenses, interest and income taxes) of $55 million (22%). I determine that it requires the employment of60 cents worth of capital (that is, net property, plant and equipment and net working capital) to produceeach dollar of sales; therefore, employed capital in Snakelight’s case comes to $150,000,000. If I thenassume that a 7.5% operating profit after inflation on employed capital could be expected from a non-branded flashlight, I deduct this generic product’s operating profit of $11,250,000 (7.5% of $150 million)in order to arrive at the profit attributable to the “Snakelight” brand name, i.e. $43,750,000. The impliedroyalty rate is therefore 17.5% ($43.75 million ÷ $250 million). In order to determine what a potential

- 21 –

licensee may be willing to pay by way of a royalty rate, I determine the brand’s strength score. Thestrength score (a maximum of 100 points) is a subjective analysis of 20 factors such as margins, marketshare growth, consumer recognition, line extension potential, transferability of brand name to otherproducts etc. In this example, if the brand strength score is determined to be 80, an arm’s length royaltyrate for Snakelight would be 14% (17.5% times 0.80). While this method has the advantage of adisciplined analytical approach to royalty rate setting, an obvious problem with its application is that moreoften than not one may encounter difficulty in obtaining a comparable generic product and relevantinformation with respect to it.

The 25% rule basically argues that the target royalty rate should result in a licensor receivingapproximately 25% of the profit from a brand name or patented product. Since substantial effort and riskis involved in final product development, manufacturing and marketing the product, the licensee should beentitled to the bulk of the profit. In my view this method on the surface has intuitive attractiveness becauseof simplicity; but it arbitrary and often difficult to apply because of several and inter-dependentassumptions.

In concluding this section on royalty rate determination, the benchmark royalty rate should be refinedbased on an assessment of the aforementioned factors affecting royalty rates and the facts obtained fromthe functional analysis. The overriding consideration in estimating a fair and appropriate royalty rate is toreview the pre-tax profit earned by the patented product. There must be adequate profits available to apotential licensee to allow payment of the royalty and still earn a reasonable profit.

The process of determining an appropriate royalty rate and that which determines the value of intellectualproperty are closely interlinked. Provided sales from the patented process or of the trademark product areidentifiable and can be reasonably estimated, one can determine the investment cost (i.e. value) if theroyalty rate is known; or conversely, the royalty rate if the lump-sum amount paid for (i.e. investment costof) the patent or trademark is known. This is expressed by the following Financial Model which usespresent value concepts to determine the appropriate royalty rate or price to be paid for the purchase of thepatent or trademark.

PV (Royalty Rate x Sales) = PV of Royalty Income = Value (of patent, trademark, brand, etc.)

It follows from the above that under the "relief from royalty" method, valuation computations generallyinclude consideration of the present value of the annual after-tax stream of royalty revenue - the result ofnot having to pay royalties - and of the present value of future income tax savings resulting from claimingcapital cost allowances (i.e. the tax shield) on the whole or specified portion of the cost of the intellectualproperty as the case may be.

To determine the present value of this after-tax stream of future royalty revenue, one requires annualestimates of revenues of the patented or trademark product estimation of the remaining economic life, theroyalty rate, the company's tax rate and a reasonable rate of return, normally based on the risk involved inrealizing the projected revenue adequate to warrant the stipulated royalty rate over the remaining economic

- 22 –

life. The risk associated with the amortization of the intellectual property (i.e. in present valuing the taxshield) is equivalent to the likelihood that the company's operations will earn just sufficient profit to claimthe amortization expense for tax purposes. The latter risk is generally lower, that is, a lower discount rateis justified, than the risk associated with generating after-tax returns to the equity holder(s).

Although relief from royalty is generally the method used to value licensable intellectual property, it is notthe only one. Another method could be capitalization of the profit margin differential between say, a brandname product and a generic product (revenue enhancement), or a patented process and an older method ofproduction (cost savings). The underlying assumption here is that the subject business owning thetrademark product or patented process has financial results comparable to others in the industry, so that theprofit differential is wholly attributable to the trademark or patent. This is seldom the case in realitybecause variances in financial results do exist and may arise due to differences in management quality,differences in distribution systems, differences in cost of capital, degree of business experience andmaturity (e.g. a mature company may have different results than start-ups), etc. However, this method canand probably should still be used as support for one's ultimate value conclusion.

Economic contributions of intellectual property - a market perspective

Corporate investments must typically pass threshold rates in order to be considered viable opportunities. Since debt and equity funds are used to finance these investments, the return that is provided must beadequate to satisfy the interest cost on debt and provide a fair rate of return on the equity funds. Thethreshold rate is the weighted average cost of capital ("WACC") in order to earn a fair rate of return oninvested capital.

Invested capital is the sum of the fair market value of equity funds and debt obligations. The cost to thecompany of the invested capital equals the rate of return that the investors expect to receive less any taxbenefits that the company enjoys, such as the deductibility for income tax purposes of interest expense ondebt.

The total fair market value of debt obligations (bonds, notes, subordinated debentures) and the variousequity components (preferred shares, common shares, warrants etc.) represent the total invested capital ofthe business enterprise. These are the funds used to obtain the complementary assets of a businessincluding land, buildings and plant, net working capital and intellectual property.

In the case of a public company it is possible to ascertain a royalty rate implied by the stock market for abundle of intellectual property assets. This is because the stock market provides the critical starting inputto this exercise by establishing the market capitalization of the company's common stock (i.e. market priceper share multiplied by the common shares outstanding).

Let us take The Gillette Company ("Gillette") as an example and see what is the market's perspective of theeconomic contributions to Gillette's intellectual property viewed as a bundle and the royalty rate impliedtherein.

- 23 –

This illustrative case study is based on Gillette’s 1998 (Dec. 31 year-end) results reported at or around themiddle of March, 1999. Its recent stock price of US$62-1/16 per share was at the high end of the 52-weekprice range of US$35-5/16 to US$62-21/32. For purposes of this case study, I have used the mid-pointprice of US$49 in order to smooth out any short-term volatility in its stock market price. Gillette recordedanother year of good progress in 1998. It launched several new products, including the phenomenallysuccessful MACH 3 shaving system, and reorganized itself for the next decade. A variety of externalevents combined to hold sales and profits at essentially 1997 levels. Gillette, however, expects earningsper share to return to the company’s typical 15% to 20% growth rate in the second half of 1999, as itsinternational business (particularly, in Asia, Russia and Latin America) bottoms out.

Gillette manufactures male and female grooming products, writing instruments and correction products,tooth brushes and oral care appliances, and alkaline batteries. The Company’s products include blades,razors, shaving preparations, and hair epilation devices, among others. The majority of its products are soldunder internationally recognized brand names such as BRAUN, PARKER PEN, WATERMAN, LIQUIDPAPER, ORAL B, DURACEL and, of course, its flagship brand - GILLETTE. In 1998, Gillette-brandedproducts accounted for 42% of sales and 52% of total profit from operations before corporate expenses. During that year, the Company spent US$1,778 million (or 17.7% of sales) on advertising and salespromotion in support of the continued customer recognition of Gillette’s brand name portfolio. In 1997and 1996, advertising and promotion expenses were 16.9% and 17.4% of respective year’s sales.

Technology is not ignored at Gillette either. Innovative technology has been the centre piece of Gillette’sstrategy and this technological superiority enables the Company to charge a premium price, some 10% to20% above the competition, in turn creating higher margins. Gillette introduced the revolutionarySENSOR line of razors in early 1990 which reportedly required US$300 million in R&D. In keeping withGillette's strategy of upgrading the type of shavers people use, in order to widen its profit margin Gilletteintroduced the SENSOREXCEL shaving system in late 1993, and in November 1995, launched itsbreakthrough SENSOREXCEL For WOMEN refillable shaving system, “the most advanced wet shavingsystem available to women”. Gillette again strengthened its clear global leadership in blades and razors, itsprincipal line of business, when in mid-1998 it launched the MACH 3 shaving system in North America,with plans for a roll-out of the product to all major markets in 1999. Developed and marketed at a cost ofUS$1 billion, the MACH 3 system has been a phenomenal success, and helped lift Gillette’s share of theblade market to its highest level in 40 years.

Gillette obviously has nurtured and supported world-class trademarks. In 1998, for instance, 47% ofGillette’s sales came from products launched in the past five years. What follows next is an attempt toshow how the trademarks pay off by isolating the economic returns that are contributed to the enterprise.

- 24 –

The Intellectual Property of Gillette:

Based on a market price of US$49 per share, Gillette had a total value of invested capital of aboutUS$58.53 billion comprising US$54.44 billion of equity and US$4.09 billion of debt and long-termobligations (both equity and debt valued at market). The Company’s various asset categories which mustequal the value of total invested capital are shown below:

Gillette Company Asset Values (US$ millions)

Value Total

Working CapitalFixed/other assetsIntangible assetsIntellectual property

$ 2,8505,1315,854

44,700

4.9%8.8%

10.0%76.3%

Total invested capital $58,535 100.0%

The source of values for the Gillette asset categories is presented below:

Source of Value

Asset Category Source of Value

Working Capital

Fixed assets

Intangible assets

Intellectual property

Book value reported in the company'saccounts.

In the absence of independent appraisals,the average of the gross and net amountsshown in the financial statements. [Netbook value alone may not be reflective ofcurrent market value.]

Estimated at 10% (1) of total investedcapital.

Residual amount of invested capital.

- 25 –

(1)Comprises trained work force, established distribution network, management information software, corporate practices

and procedures and organizational goodwill. The % is estimated based on experience from several years of consulting in

M&A.

The economic contribution from the Gillette intellectual property requires an allocation of the total debt-free net income ("DFNI") of the enterprise. Based on the value of the different assets used in the businessand the relative investment risk associated with each, the intellectual property contribution can be isolated.

The weighted average cost of capital ("WACC") for Gillette is the key input. The WACC is based uponthe market value of equity and the value of long-term debt. WACC represents the minimum amount ofinvestment return that is considered acceptable from operating the business. When the cost of these capitalcomponents are weighted by their percentage of the total invested capital, a WACC of 10.6% for Gillette isthe result (based on 1998 financial information).

The WACC requirement can be allocated among the assets that are employed within the businessenterprise. The allocation is conducted with respect to the amount of investment risk that each componentrepresents to the business enterprise.

Appropriate Return on Monetary Assets:

Monetary assets of a business are its net working capital. Working capital is considered the most liquidasset of a business. Given the relative liquidity of working capital, the amount of investment risk isinherently low in comparison to that of the other asset categories. An appropriate rate of return to associatewith the working capital component is typically lower than the overall WACC. A surrogate rate of returncan be used: that which is available from investment in short-term securities of low risk levels. The rateavailable on say 90-day U.S. certificates of deposits, currently 4.72%, can serve as a benchmark,concluding at 5.0% for purposes of this illustration.

Appropriate Return on Tangible Assets:

Tangible assets or fixed assets comprise production machinery, equipment, transportation fleet, officebuildings, land etc. While these assets are not as liquid as working capital, they still possess someelements of marketability. They can often be sold to other companies or used for alternate businesspurposes. This marketability allows a partial return for the investment in fixed assets of the businessshould the business fail.

An indication of the rate of return that is contributed by these assets can be pegged at about the interest rateat which commercial banks make loans using fixed assets as collateral, 7.75% for purposes of thisillustration. This rate reflects the higher risk undertaken by equity owners than by lenders.Appropriate Return on Intellectual Property:

Intangible assets and intellectual property are considered the most risky asset components of the overall

- 26 –

business enterprise. Trademarks can fall out of favour with the attitudes of society, and patents canbecome obsolete by advancing technology of competitors. This increases their risk. A higher rate ofreturn on these assets is therefore required. Since the overall return of the enterprise is established as theweighted average cost of capital, and since reasonable returns for the monetary and tangible assets can beestimated, one is then in a position to derive an appropriate rate of return earned by intangible assets andintellectual property.

For Gillette, the overall minimum acceptable return was determined to be 10.6%. Based upon the relativerisk discussion presented earlier, the following table assigns different levels of required return to thedifferent asset categories.

Allocation of Required Rate of Return toDifferent Asset Categories (US$MM's)

Asset Category AssetValue

% ofTotal

RequiredReturn(%)

WeightedReturn(%)

% ofWeightedReturn

AllocationOfDFNI

Working CapitalFixed assetsIntangiblesIntellectual

property

$ 2,8505,1315,854

44,700

4.98.810.0

76.3

0.250.681.12

8.55

2.366.4110.57

80.66

$ 26.973.1120.5

$919.5

$58,535 100%

5.007.7511.21

11.21

10.60% 100% $1,140.0

As a result of these investment rate of return requirements, Gillette's intellectual property accounts for justover 80% of the total debt-free net income (“DFNI”) of the company. Of the total DFNI of US$1,140million generated by Gillette in 1998, US$919.5 million is attributed to the intellectual property. Withoutit, the company would have earned only US$220.5 million. Therefore, the employment of intellectualproperty enabled Gillette to earn excess returns.

As a percent of total 1998 sales of U.S.$10,956 million, the excess return of US$919.5 million representsabout 9.1%. This is on an after-tax basis. Since royalties are expressed in pre-tax terms, the pre-taxindication of royalties on Gillette's intellectual property viewed as a bundle is 14% [91÷(1-0.35)=14.0%]as indicated by the marketplace.

One could go through a similar exercise for various public companies within selected industries, andobtain a reasonably good approximation of the market's perception of current royalty rates in that industry.

- 27 –

Conclusion

If the saying "valuation is an art and not a science" is a truism, it is never more so than in the case ofintellectual property. This is because of the nature of the asset itself. Not only is the valuator faced withthe initial task of establishing the very existence of intellectual property, he must next determine whether itis valuable (that is, is it contributing to the earning power of the business) and commercially transferable,and finally, quantify such value.

In this paper, I have discussed the relief from royalty method of valuing intellectual property and shownhow the market's perspective of economic contributions among complementary assets of a public companycan guide one to determine a reasonable royalty rate on its bundle of intellectual property.

When valuing intellectual property, no one method can be sufficiently all conclusive -- the key is to useseveral methods to support one's initial findings. Finally, while knowledge (of the various techniques andconcepts) is critical, one cannot overemphasize the importance of a disciplined, logical approach and agood dose of common sense and reasoned judgement in arriving at reasonable royalty rates and hence, asupportable valuation.

*****

References:

• “Music bonds are failing to prove chart toppers”, Financial Times, March 2, 1999.

• “Valuation of Intellectual Property Assets: The Foundation for Risk Management and Financing”, apaper presented by Zareer Pavri of PricewaterhouseCoopers, at the INFONEX Conference: “BusinessStrategies for Managing Intellectual Property: A Practical Course”, September 23 and 29, 1997.

• “Valuation of Intellectual Property: The Legal and Business Issues”, a paper presented by Zareer Pavriof PricewaterhouseCoopers, at the Canadian Institute Conference: “Licensing Intellectual Property:The Legal and Business Issues”, January 23 and 24, 1996.

• "Determining an Arm's Length Price: A Case Study for Intangibles", a paper presented by A. J.d'Ombrain of PricewaterhouseCoopers, at the February 1994 Canadian Institute seminar on TransferPricing.

• "Where the Value in a Trademark Lies", Zareer Pavri of PricewaterhouseCoopers, CA Magazine,February 1987.

• "Establishing Fair Market Value of Intangible Assets" by Paul Gross of The American AppraisalCompany, appearing in The Journal of Business Valuation, July 1977.

• "Eureka! Now What? An Introduction to Patents, Trade Marks and Copyright", Ronald Dimock,

- 28 –

Donald Cameron and Brenda Boardman, published by CCH Canadian Limited.

• "Valuation of Intellectual Property and Intangible Assets" by Gordon Smith and Russell Parr,published by John Wiley & Sons.

• "Tax Management Special Report: Transfer Pricing", January 6, 1993; and appearing therein,"Unrelated-Party Licensing Practices and Factors Affecting Royalty Rates: Results of a Survey", byDaniel McGavock of IPC Group Inc.

© 1999 PricewaterhouseCoopers LLP. PricewaterhouseCoopers LLP refers to the Canadian firm of PricewaterhouseCoopers and other members of the worldwidePricewaterhouseCoopers organization.